You can subscribe to this list here.
| 2001 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2002 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
(1) |
Sep
|
Oct
|
Nov
(1) |
Dec
|
| 2003 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
(1) |
Sep
|
Oct
(83) |
Nov
(57) |
Dec
(111) |
| 2004 |
Jan
(38) |
Feb
(121) |
Mar
(107) |
Apr
(241) |
May
(102) |
Jun
(190) |
Jul
(239) |
Aug
(158) |
Sep
(184) |
Oct
(193) |
Nov
(47) |
Dec
(68) |
| 2005 |
Jan
(190) |
Feb
(105) |
Mar
(99) |
Apr
(65) |
May
(92) |
Jun
(250) |
Jul
(197) |
Aug
(128) |
Sep
(101) |
Oct
(183) |
Nov
(186) |
Dec
(42) |
| 2006 |
Jan
(102) |
Feb
(122) |
Mar
(154) |
Apr
(196) |
May
(181) |
Jun
(281) |
Jul
(310) |
Aug
(198) |
Sep
(145) |
Oct
(188) |
Nov
(134) |
Dec
(90) |
| 2007 |
Jan
(134) |
Feb
(181) |
Mar
(157) |
Apr
(57) |
May
(81) |
Jun
(204) |
Jul
(60) |
Aug
(37) |
Sep
(17) |
Oct
(90) |
Nov
(122) |
Dec
(72) |
| 2008 |
Jan
(130) |
Feb
(108) |
Mar
(160) |
Apr
(38) |
May
(83) |
Jun
(42) |
Jul
(75) |
Aug
(16) |
Sep
(71) |
Oct
(57) |
Nov
(59) |
Dec
(152) |
| 2009 |
Jan
(73) |
Feb
(213) |
Mar
(67) |
Apr
(40) |
May
(46) |
Jun
(82) |
Jul
(73) |
Aug
(57) |
Sep
(108) |
Oct
(36) |
Nov
(153) |
Dec
(77) |
| 2010 |
Jan
(42) |
Feb
(171) |
Mar
(150) |
Apr
(6) |
May
(22) |
Jun
(34) |
Jul
(31) |
Aug
(38) |
Sep
(32) |
Oct
(59) |
Nov
(13) |
Dec
(62) |
| 2011 |
Jan
(114) |
Feb
(139) |
Mar
(126) |
Apr
(51) |
May
(53) |
Jun
(29) |
Jul
(41) |
Aug
(29) |
Sep
(35) |
Oct
(87) |
Nov
(42) |
Dec
(20) |
| 2012 |
Jan
(111) |
Feb
(66) |
Mar
(35) |
Apr
(59) |
May
(71) |
Jun
(32) |
Jul
(11) |
Aug
(48) |
Sep
(60) |
Oct
(87) |
Nov
(16) |
Dec
(38) |
| 2013 |
Jan
(5) |
Feb
(19) |
Mar
(41) |
Apr
(47) |
May
(14) |
Jun
(32) |
Jul
(18) |
Aug
(68) |
Sep
(9) |
Oct
(42) |
Nov
(12) |
Dec
(10) |
| 2014 |
Jan
(14) |
Feb
(139) |
Mar
(137) |
Apr
(66) |
May
(72) |
Jun
(142) |
Jul
(70) |
Aug
(31) |
Sep
(39) |
Oct
(98) |
Nov
(133) |
Dec
(44) |
| 2015 |
Jan
(70) |
Feb
(27) |
Mar
(36) |
Apr
(11) |
May
(15) |
Jun
(70) |
Jul
(30) |
Aug
(63) |
Sep
(18) |
Oct
(15) |
Nov
(42) |
Dec
(29) |
| 2016 |
Jan
(37) |
Feb
(48) |
Mar
(59) |
Apr
(28) |
May
(30) |
Jun
(43) |
Jul
(47) |
Aug
(14) |
Sep
(21) |
Oct
(26) |
Nov
(10) |
Dec
(2) |
| 2017 |
Jan
(26) |
Feb
(27) |
Mar
(44) |
Apr
(11) |
May
(32) |
Jun
(28) |
Jul
(75) |
Aug
(45) |
Sep
(35) |
Oct
(285) |
Nov
(99) |
Dec
(16) |
| 2018 |
Jan
(8) |
Feb
(8) |
Mar
(42) |
Apr
(35) |
May
(23) |
Jun
(12) |
Jul
(16) |
Aug
(11) |
Sep
(8) |
Oct
(16) |
Nov
(5) |
Dec
(8) |
| 2019 |
Jan
(9) |
Feb
(28) |
Mar
(4) |
Apr
(10) |
May
(7) |
Jun
(4) |
Jul
(4) |
Aug
|
Sep
(4) |
Oct
|
Nov
(23) |
Dec
(3) |
| 2020 |
Jan
(19) |
Feb
(3) |
Mar
(22) |
Apr
(17) |
May
(10) |
Jun
(69) |
Jul
(18) |
Aug
(23) |
Sep
(25) |
Oct
(11) |
Nov
(20) |
Dec
(9) |
| 2021 |
Jan
(1) |
Feb
(7) |
Mar
(9) |
Apr
|
May
(1) |
Jun
(8) |
Jul
(6) |
Aug
(8) |
Sep
(7) |
Oct
|
Nov
(2) |
Dec
(23) |
| 2022 |
Jan
(23) |
Feb
(9) |
Mar
(9) |
Apr
|
May
(8) |
Jun
(1) |
Jul
(6) |
Aug
(8) |
Sep
(30) |
Oct
(5) |
Nov
(4) |
Dec
(6) |
| 2023 |
Jan
(2) |
Feb
(5) |
Mar
(7) |
Apr
(3) |
May
(8) |
Jun
(45) |
Jul
(8) |
Aug
|
Sep
(2) |
Oct
(14) |
Nov
(7) |
Dec
(2) |
| 2024 |
Jan
(4) |
Feb
(4) |
Mar
|
Apr
(7) |
May
(2) |
Jun
(1) |
Jul
|
Aug
(5) |
Sep
|
Oct
|
Nov
(4) |
Dec
(14) |
| 2025 |
Jan
(22) |
Feb
(6) |
Mar
(5) |
Apr
(14) |
May
(6) |
Jun
(11) |
Jul
(19) |
Aug
|
Sep
(17) |
Oct
(1) |
Nov
(2) |
Dec
(18) |
| 2026 |
Jan
|
Feb
|
Mar
(5) |
Apr
|
May
(2) |
Jun
(1) |
Jul
(6) |
Aug
(1) |
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Werner S. <sm...@ia...> - 2010-02-25 22:03:45
|
Hi, after some messing around I finally made the wxt(p) driver work. What I did was: * Download the GTK_2.18.5-X11.pkg package from http://r.research.att.com/ and install it * Download the wxWidgets 2.8.11-rc2 source and compile a static build Compiling gnuplot 4.4/4.5 works fine, pngcairo/pdfcairo work, but the wxt terminal hangs. Looking at the code this is because wxt uses a second thread which calls wxApp::OnRun which shouldn't be done, and obviously doesn't work in Mac OS X. I patched gnuplot 4.4 from cvs with the patch adding another terminal wxtp which works similar to gnuplot_x11. The patch didn't apply cleanly, but this could be sorted out. It compiles and runs, but window doesn't get focus - known problem, since gnuplot is not an app, but with a little "hack" using CPSEnableForegroundOperation() this can also be solved and I got a working (sort of) wxtp terminal outside of an app bundle. I just want to known what the status of wxt is now. Is wxtp supposed to be applied to cvs? The patch is over a year old. Anyway, just want to know where this is heading, maybe I can help a little (having some experience being the maintainer of the wxWidgets driver of PLplot). Regards, Werner -- Dr. Werner Smekal Institut fuer Angewandte Physik Technische Universitaet Wien Wiedner Hauptstr 8-10/134 A-1040 Wien Austria DVR-Nr: 0005886 email: sm...@ia... (GPG: EDCAF4A79) web: http://www.iap.tuwien.ac.at/~smekal phone: +43-(0)1-58801-13463 (office) +43-(0)1-58801-13469 (laboratory) fax: +43-(0)1-58801-13499 |
|
From: Hans-Bernhard B. <HBB...@t-...> - 2010-02-25 21:04:36
|
Ethan Merritt wrote: > On Tuesday 23 February 2010 02:45:54 Shigeharu TAKENO wrote: >> However, we can not use the character "$" as the part of the >> variable name. > > You are right. I will make this correction in the documentation. > > I wonder why or when this changed? I don't recall when, but the "why" is pretty obvious: we use $ as an operator for 'call' arguments and as an abbreviation of column(). |
|
From: Shigeharu T. <sh...@ie...> - 2010-02-25 02:20:15
|
shige 02/25 2010 ---------------- I wrote: | I prepared sample scripts, data, and results. See | | http://takeno.iee.niit.ac.jp/~shige/unix/gnuplot/data/mkbin.c | http://takeno.iee.niit.ac.jp/~shige/unix/gnuplot/data/mkbin.gp | http://takeno.iee.niit.ac.jp/~shige/unix/gnuplot/data/mkbin1.gif | http://takeno.iee.niit.ac.jp/~shige/unix/gnuplot/data/mkbin2.gif I explane them. mkbin.c make the following binary matrix data: 3.0 5.0 6.0 7.0 (<N+1> <y0> <y1> <y2>) 0.0 1.0 2.0 3.0 ( <x0> <z0,0> <z0,1> <z0,2>) 0.1 3.0 1.0 4.0 ( <x1> <z1,0> <z1,1> <z1,2>) 0.2 2.0 2.5 3.0 ( <x2> <z2,0> <z2,1> <z2,2>) Gnuplot script mkbin.gp makes images mkbin1.gif and mkbin2.gif from the data made by mkbin.c. Gnuplot.doc says that gnuplot treat it as the following: 0.0 5.0 1.0 0.0 6.0 2.0 0.0 7.0 3.0 0.1 5.0 3.0 0.1 6.0 1.0 0.1 7.0 4.0 0.2 5.0 2.0 0.2 6.0 2.5 0.2 7.0 3.0 However, images mkbin1.gif and mkbin2.gif show that these are made from the data 5.0 0.0 1.0 6.0 0.0 2.0 7.0 0.0 3.0 5.0 0.1 3.0 6.0 0.1 1.0 7.0 0.1 4.0 5.0 0.2 2.0 6.0 0.2 2.5 7.0 0.2 3.0 Moreover, I found a problem in these tests. The following script ! ./mkbin > mkbin.dat splot "mkbin.dat" binary matrix with lines works. But, splot "< ./mkbin" binary matrix with lines does not work and gnuplot says "line 1: Data file is empty". OS: Solaris 9 (sparc) CC: gcc-3.4.3 gnuplot: gnuplot-4.5 (CVS version) +========================================================+ Shigeharu TAKENO NIigata Institute of Technology kashiwazaki,Niigata 945-1195 JAPAN sh...@ie... TEL(&FAX): +81-257-22-8161 +========================================================+ |
|
From: Shigeharu T. <sh...@ie...> - 2010-02-24 23:55:05
|
shige 02/25 2010 ---------------- I wrote: | 1) Gnuplot help says as the following (section "binary matrix"): | | which are converted into triplets: | <x0> <y0> <z0,0> | <x0> <y1> <z0,1> | <x0> <y2> <z0,2> | : : : | <x0> <yN> <z0,N> | | <x1> <y0> <z1,0> | <x1> <y1> <z1,1> | : : : | | However, gnuplot seems to treat it as the following: | | <y0> <x0> <z0,0> | <y0> <x1> <z0,1> | <y0> <x2> <z0,2> | : : : | <y0> <xN> <z0,N> | | <y1> <x0> <z1,0> | <y1> <x1> <z1,1> | : : : | | I checked it by | | splot "binary.dat" binary matrix every :::0::1 w l | | for an asymmetric binary data "binary.dat". I prepared sample scripts, data, and results. See http://takeno.iee.niit.ac.jp/~shige/unix/gnuplot/data/mkbin.c http://takeno.iee.niit.ac.jp/~shige/unix/gnuplot/data/mkbin.gp http://takeno.iee.niit.ac.jp/~shige/unix/gnuplot/data/mkbin1.gif http://takeno.iee.niit.ac.jp/~shige/unix/gnuplot/data/mkbin2.gif +========================================================+ Shigeharu TAKENO NIigata Institute of Technology kashiwazaki,Niigata 945-1195 JAPAN sh...@ie... TEL(&FAX): +81-257-22-8161 +========================================================+ |
|
From: Ethan M. <merritt@u.washington.edu> - 2010-02-24 20:56:11
|
On Tuesday 23 February 2010 02:45:54 Shigeharu TAKENO wrote: > 4) Gnuplot help says as the following (section "variables"): > > Valid names are the same as in most programming languages: they must begin > with a letter, but subsequent characters may be letters, digits, "$", or "_". > > However, we can not use the character "$" as the part of the > variable name. You are right. I will make this correction in the documentation. I wonder why or when this changed? It must have been a long time ago. This is interesting, because I think it means we could use $ as an operator if necessary. Ethan |
|
From: Ethan M. <merritt@u.washington.edu> - 2010-02-23 20:16:12
|
On Tuesday 23 February 2010 02:45:54 Shigeharu TAKENO wrote:
> 3) Gnuplot help says as the following ("set datafile missing"):
>
> Example:
> set datafile missing "?"
> set style data lines
> plot '-'
> 1 10
> 2 20
> 3 ?
> 4 40
> 5 50
> e
> ....
> The first `plot` will recognize only the first datum in the "3 ?" line. It
> will use the single-datum-on-a-line convention that the line number is "x"
> and the datum is "y", so the point will be plotted (in this case erroneously)
> at (2,3).
>
> However, it, mentioned above, is the behavior when not using 'set
> datafile missing "?"'. Using 'set datafile missing "?"', the line
> "3 ?" will be ignored correctly.
Yes. This could be stated more clearly in the documentation.
We should try to improve the wording.
Ethan
|
|
From: Ethan M. <merritt@u.washington.edu> - 2010-02-23 20:09:06
|
On Tuesday 23 February 2010 02:45:54 Shigeharu TAKENO wrote: > shige 02/23 2010 > ---------------- > > Recently, a new book for gnuplot-4.2 was published last year, > written by M.Yamamoto in Japanese. In it, he pointed out some > problems about differences between the gnuplot binary and its > help. > > 2) Help for png/jpeg/png terminal says: > > set terminal gif medium size 640,480 \ > xffffff x000000 x404040 \ > xff0000 xffa500 x66cdaa xcdb5cd \ > xadd8e6 x0000ff xdda0dd x9500d3 # defaults > > However these colors are not the defaults of the terminals. > I think the comment "#defaults" should be removed. You are correct. This entire mechanism is obsolete. In fact, the command does not seem to work as documented in the current 4.4 or 4.5 cvs source, even apart from issue of the default values. [and what did it ever mean "the x and y axis color"?? Was it ever possible to use colors for the x and y axes that were different from the border color?] I suggest that we should remove this section from the documentation. Instead we will cross-reference the sections explaining that colors can now be specified for all terminals using commands such as plot ... lc rgbcolor "#RRGGBB" set border lc rgbcolor "#RRGGBB" or set linetype N lc rgbcolor '#RRGGBB" The png/gif/jpeg terminals can continue to accept this terminal option for backwards compatibility with old scripts even though it is no longer recommended, documented, or supported. However, interpretation of the very first color as the background color must remain. There is currently no other way to set the background color correctly for the png and jpeg terminals. It would be reasonable to add new command "set background", but we don't have such a thing now. Ethan |
|
From: Shigeharu T. <sh...@ie...> - 2010-02-23 10:46:08
|
shige 02/23 2010
----------------
Recently, a new book for gnuplot-4.2 was published last year,
written by M.Yamamoto in Japanese. In it, he pointed out some
problems about differences between the gnuplot binary and its
help.
1) Gnuplot help says as the following (section "binary matrix"):
which are converted into triplets:
<x0> <y0> <z0,0>
<x0> <y1> <z0,1>
<x0> <y2> <z0,2>
: : :
<x0> <yN> <z0,N>
<x1> <y0> <z1,0>
<x1> <y1> <z1,1>
: : :
However, gnuplot seems to treat it as the following:
<y0> <x0> <z0,0>
<y0> <x1> <z0,1>
<y0> <x2> <z0,2>
: : :
<y0> <xN> <z0,N>
<y1> <x0> <z1,0>
<y1> <x1> <z1,1>
: : :
I checked it by
splot "binary.dat" binary matrix every :::0::1 w l
for an asymmetric binary data "binary.dat".
2) Help for png/jpeg/png terminal says:
set terminal gif medium size 640,480 \
xffffff x000000 x404040 \
xff0000 xffa500 x66cdaa xcdb5cd \
xadd8e6 x0000ff xdda0dd x9500d3 # defaults
However these colors are not the defaults of the terminals.
I think the comment "#defaults" should be removed.
3) Gnuplot help says as the following ("set datafile missing"):
Example:
set datafile missing "?"
set style data lines
plot '-'
1 10
2 20
3 ?
4 40
5 50
e
....
The first `plot` will recognize only the first datum in the "3 ?" line. It
will use the single-datum-on-a-line convention that the line number is "x"
and the datum is "y", so the point will be plotted (in this case erroneously)
at (2,3).
However, it, mentioned above, is the behavior when not using 'set
datafile missing "?"'. Using 'set datafile missing "?"', the line
"3 ?" will be ignored correctly.
4) Gnuplot help says as the following (section "variables"):
Valid names are the same as in most programming languages: they must begin
with a letter, but subsequent characters may be letters, digits, "$", or "_".
However, we can not use the character "$" as the part of the
variable name.
+========================================================+
Shigeharu TAKENO NIigata Institute of Technology
kashiwazaki,Niigata 945-1195 JAPAN
sh...@ie... TEL(&FAX): +81-257-22-8161
+========================================================+
|
|
From: Tatsuro M. <tma...@ya...> - 2010-02-23 06:24:44
|
Hello This is just a record of my building NetBSD editline library under the cygwin. This might be useful to those who distribute gnuplot binaries with the readline/editline feature. The NetBSD editline library is not distributed the cygwin official site. One has to build from source. Download site http://thrysoee.dk/editline/ Patch is required to build it under the cygwin. It is available from the below http://old.nabble.com/R%3A-Error-when-compiling-mySQL-client-libraries-p27339363.html Regards Tatsuro --- Tatsuro MATSUOKA wrote: > Hello > > The cygwin binary distributions of gnuplot 4.5. and 4.4rc1cvs are updated > > http://www.tatsuromatsuoka.com/gnuplot/Eng/cygbin/ > > ================================================================ > I decide to use the NetBSD editline library instead of the GNU readline > for my binary distribution to avoid the possibility of violation of the GPL licence of > the GNU readline. Please refer the below > http://old.nabble.com/Re%3A-readline-to-gnuplot.exe-(console-mode)-for-windows-p27677233.html > > Note that when user build binaries of gnuplot from source by themselves, > the linking with the GNU readline is not a problem. > > ======================================================= > The default location of the x11 resource file on the current cygwin > is /etc/defaults/etc/X11/app-defaults/. If you do not find the file 'Gnuplot' > in /etc/defaults/etc/X11/app-defaults/, please copy it from > gp45-winbinX11/X11/app-defaults in the zip file paste it there. > Please refer > http://old.nabble.com/Re%3A-inconsistency-between-4.4-cvs-and-4.5-for-default-install-directory-of-x11-resources-(Gnuplot)-p27672453.html > > Regards > > Tatsuro > > -------------------------------------- > VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] > http://pr.mail.yahoo.co.jp/olympic/ > > ------------------------------------------------------------------------------ > Download Intel¢î Parallel Studio Eval > Try the new software tools for yourself. Speed compiling, find bugs > proactively, and fine-tune applications for parallel performance. > See why Intel Parallel Studio got high marks during beta. > http://p.sf.net/sfu/intel-sw-dev > _______________________________________________ > gnuplot-beta mailing list > gnu...@li... > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta > -------------------------------------- VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] http://pr.mail.yahoo.co.jp/olympic/ |
|
From: Ethan M. <merritt@u.washington.edu> - 2010-02-23 03:08:55
|
On Monday 22 February 2010 17:38:10 Tatsuro MATSUOKA wrote:
> Hello
>
> Thanks for the fix.
>
> I have confirmed the fix on the gnuplot 4.5 (CVS).
>
> 2010-02-22 Don Taber <dt...@to...>
>
> * src/win/wgraph.c (SelFont): After a font change, call
> do_string_replot("") so that it is immediately visible.
>
> On the 4.4cvs, the above fix cannot be found. I would like to just confirm that whether the patch is
> applied only to the 4.5.
Yes. I was reluctant to put it in the 4.4 cvs so close to release date.
But if people think it is important, and won't break anything, I can add it.
Ethan
>
> Regards
>
> Tatsuro
>
> --- don taber wrote:
>
> > > > One problem I note in both cases: Font changes made with the "Choose
> > > > Font" Windows dialog do not take effect unless followed by a replot
> >
> > > > There is something missing from the code that redraws the plot when
> > > > the dialog is closed. I'm sure that someone who is more of a Windows
> > > > programmer than I can fix this easily.
> > >
> > > No, there is nothing missing in the code that redraws the plot.
> > > And I'm not sure this can be easily fixed.
> > >
> > > This is a manifestation of a principal problem in gnuplot's plotting
> > > algorithm.
> > >
> >
> > Benjamin,
> >
> > Thanks for the explanation of why the current code needs
> > the replot. This analysis seems to suggest that my (revised) patch
> > calling do_string_replot(), as suggested by Ethan, is
> > the best solution.
> >
> > Don Taber
|
|
From: Tatsuro M. <tma...@ya...> - 2010-02-23 02:46:44
|
Hello --- Ethan Merritt wrote: > > On the 4.4cvs, the above fix cannot be found. I would like to just confirm that whether the > patch is > > applied only to the 4.5. > > Yes. I was reluctant to put it in the 4.4 cvs so close to release date. > But if people think it is important, and won't break anything, I can add it. > > Ethan OK. I agree with your judgment. Regards Tatsuro -------------------------------------- VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] http://pr.mail.yahoo.co.jp/olympic/ |
|
From: Tatsuro M. <tma...@ya...> - 2010-02-23 02:38:21
|
Hello
Thanks for the fix.
I have confirmed the fix on the gnuplot 4.5 (CVS).
2010-02-22 Don Taber <dt...@to...>
* src/win/wgraph.c (SelFont): After a font change, call
do_string_replot("") so that it is immediately visible.
On the 4.4cvs, the above fix cannot be found. I would like to just confirm that whether the patch is
applied only to the 4.5.
Regards
Tatsuro
--- don taber wrote:
> > > One problem I note in both cases: Font changes made with the "Choose
> > > Font" Windows dialog do not take effect unless followed by a replot
>
> > > There is something missing from the code that redraws the plot when
> > > the dialog is closed. I'm sure that someone who is more of a Windows
> > > programmer than I can fix this easily.
> >
> > No, there is nothing missing in the code that redraws the plot.
> > And I'm not sure this can be easily fixed.
> >
> > This is a manifestation of a principal problem in gnuplot's plotting
> > algorithm.
> >
>
> Benjamin,
>
> Thanks for the explanation of why the current code needs
> the replot. This analysis seems to suggest that my (revised) patch
> calling do_string_replot(), as suggested by Ethan, is
> the best solution.
>
> Don Taber
>
--------------------------------------
VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
http://pr.mail.yahoo.co.jp/olympic/
|
|
From: don t. <dt...@to...> - 2010-02-22 22:38:31
|
> > One problem I note in both cases: Font changes made with the "Choose > > Font" Windows dialog do not take effect unless followed by a replot > > There is something missing from the code that redraws the plot when > > the dialog is closed. I'm sure that someone who is more of a Windows > > programmer than I can fix this easily. > > No, there is nothing missing in the code that redraws the plot. > And I'm not sure this can be easily fixed. > > This is a manifestation of a principal problem in gnuplot's plotting > algorithm. > Benjamin, Thanks for the explanation of why the current code needs the replot. This analysis seems to suggest that my (revised) patch calling do_string_replot(), as suggested by Ethan, is the best solution. Don Taber |
|
From: Ethan M. <merritt@u.washington.edu> - 2010-02-22 22:24:23
|
On Monday 22 February 2010 12:35:01 don taber wrote:
> Thanks for the tip. Revised patch below works with in-line data.
>
> Don Taber
Applied to CVS. Thanks.
But this is the sort of last-minute change that I worry has a possibility
of breaking something. So I will hold off applying it to 4.4 until after
the release of patchlevel 0.
Unless people think this is important enough to put in the 4.4.0 release
even though it is not widely untested?
Ethan
> ---------------------------------------------------
>
> --- wgraph.c.orig 2010-02-19 11:17:11.000000000 -0800
> +++ wgraph.c 2010-02-22 12:26:56.000000000 -0800
> @@ -771,7 +771,6 @@
> }
> return;
> }
> -
> static void
> SelFont(LPGW lpgw)
> {
> @@ -815,6 +814,8 @@
> strcpy(lpgw->deffontname,lpgw->fontname);
> lpgw->deffontsize = lpgw->fontsize;
> SendMessage(lpgw->hWndGraph,WM_COMMAND,M_REBUILDTOOLS,0L);
> + /* DBT 2010-02-22 replot to force immediate font change, volatile data OK */
> + do_string_replot("");
> }
> #endif
> }
|
|
From: don t. <dt...@to...> - 2010-02-22 20:35:15
|
> > > One problem I note in both cases: Font changes made with the "Choose
> > > Font" Windows dialog do not take effect unless followed by a replot
> >
> > The appended patch fixes the problem for me. I general, I think
> > it should be okay because you cannot get to that dialog unless
>
> Careful. It may not work for volatile input data, e.g.
> in-line data typed from the command line.
>
> It would probably be better to set up a call to do_string_replot()
> rather than to call replotrequest() directly. The routine tests
Thanks for the tip. Revised patch below works with in-line data.
Don Taber
---------------------------------------------------
--- wgraph.c.orig 2010-02-19 11:17:11.000000000 -0800
+++ wgraph.c 2010-02-22 12:26:56.000000000 -0800
@@ -771,7 +771,6 @@
}
return;
}
-
static void
SelFont(LPGW lpgw)
{
@@ -815,6 +814,8 @@
strcpy(lpgw->deffontname,lpgw->fontname);
lpgw->deffontsize = lpgw->fontsize;
SendMessage(lpgw->hWndGraph,WM_COMMAND,M_REBUILDTOOLS,0L);
+ /* DBT 2010-02-22 replot to force immediate font change, volatile data OK */
+ do_string_replot("");
}
#endif
}
|
|
From: Ethan M. <merritt@u.washington.edu> - 2010-02-22 17:52:32
|
On Monday 22 February 2010 08:58:46 don taber wrote: > > "don taber" wrote : > > ====================================================================== > > One problem I note in both cases: Font changes made with the "Choose > > Font" Windows dialog do not take effect unless followed by a replot > ... > > > > Any suggestions? > > The appended patch fixes the problem for me. I general, I think > it should be okay because you cannot get to that dialog unless > you have already successfully performed a plot. So replotting > should succeed. Careful. It may not work for volatile input data, e.g. in-line data typed from the command line. > The only abberant case I have found is to > splot a 3d surface and then issue a bad 2d plot command. > Now replotting fails because (I think) the 'is_3d_plot' variable got > changed by the bad plot command. > > Don Taber It would probably be better to set up a call to do_string_replot() rather than to call replotrequest() directly. The routine tests for the presence of volatile data and then calls either refresh_request() or replotrequest() as appropriate. There may well be other places that the same care should be taken. I am not sure the windows terminal driver was ever audited for this when the refresh/replot distinction was introduced. (working off pure logic here, I haven't tested the code) Ethan > ------------------------------------------------------------ > > > --- wgraph.c.orig 2010-02-19 11:17:11.000000000 -0800 > +++ wgraph.c 2010-02-19 11:17:37.000000000 -0800 > @@ -815,6 +815,7 @@ > strcpy(lpgw->deffontname,lpgw->fontname); > lpgw->deffontsize = lpgw->fontsize; > SendMessage(lpgw->hWndGraph,WM_COMMAND,M_REBUILDTOOLS,0L); > + replotrequest(); > } > #endif > } > > ------------------------------------------------------------------------------ |
|
From: don t. <dt...@to...> - 2010-02-22 17:38:24
|
> "don taber" wrote : > ====================================================================== > One problem I note in both cases: Font changes made with the "Choose > Font" Windows dialog do not take effect unless followed by a replot ... > > Any suggestions? The appended patch fixes the problem for me. I general, I think it should be okay because you cannot get to that dialog unless you have already successfully performed a plot. So replotting should succeed. The only abberant case I have found is to splot a 3d surface and then issue a bad 2d plot command. Now replotting fails because (I think) the 'is_3d_plot' variable got changed by the bad plot command. Don Taber ------------------------------------------------------------ --- wgraph.c.orig 2010-02-19 11:17:11.000000000 -0800 +++ wgraph.c 2010-02-19 11:17:37.000000000 -0800 @@ -815,6 +815,7 @@ strcpy(lpgw->deffontname,lpgw->fontname); lpgw->deffontsize = lpgw->fontsize; SendMessage(lpgw->hWndGraph,WM_COMMAND,M_REBUILDTOOLS,0L); + replotrequest(); } #endif } |
|
From: Benjamin L. <lin...@gm...> - 2010-02-22 17:24:59
|
> One problem I note in both cases: Font changes made with the "Choose
> Font" Windows dialog do not take effect unless followed by a replot
> command. The plot is immediately redrawn when the dialog is dismissed
> with 'Okay', but the font does not change and if the Choose Font
> dialog is called again without an intervening replot, the old font is
> displayed. This separate replot command should not be necessary.
> There is something missing from the code that redraws the plot when
> the dialog is closed. I'm sure that someone who is more of a Windows
> programmer than I can fix this easily.
No, there is nothing missing in the code that redraws the plot.
And I'm not sure this can be easily fixed.
This is a manifestation of a principal problem in gnuplot's plotting
algorithm.
One of the very first commands (actually when I debugged this problem,
*the* first command) sent to the terminal when creating a plot is a
reset-font-to-the-default-font command.
Calling stack:
do_plot() (graphics.c)
boundary() (graphics.c)
find_maxl_keys() (graphics.c)
ignore_enhanced() (term.c)
term->set_font("")
This command will contain the default font at startup. Fine.
Now you change the default font from the dialog box. This changes the
default font name for the terminal. But redrawing the graph simply
re-executes the queued plotting commands. And these do not change unless
you execute a "replot".
Which means that the font hasn't really changed, until you issue a replot.
The reason this works for the line styles is that the line style itself
is not hard-coded into the command queue. It just queues the style
number to use, not the style definition (so if you change the style
definition and re-execute the commands you'll get an updated display).
However, the font names are hard-coded. So there is no special command
for "set to the default font". It's just "set to font FOO" (which
happens to be the default font at the time the command is queued).
benjamin
|
|
From: Tatsuro M. <tma...@ya...> - 2010-02-22 07:02:35
|
Hello The cygwin binary distributions of gnuplot 4.5. and 4.4rc1cvs are updated http://www.tatsuromatsuoka.com/gnuplot/Eng/cygbin/ ================================================================ I decide to use the NetBSD editline library instead of the GNU readline for my binary distribution to avoid the possibility of violation of the GPL licence of the GNU readline. Please refer the below http://old.nabble.com/Re%3A-readline-to-gnuplot.exe-(console-mode)-for-windows-p27677233.html Note that when user build binaries of gnuplot from source by themselves, the linking with the GNU readline is not a problem. ======================================================= The default location of the x11 resource file on the current cygwin is /etc/defaults/etc/X11/app-defaults/. If you do not find the file 'Gnuplot' in /etc/defaults/etc/X11/app-defaults/, please copy it from gp45-winbinX11/X11/app-defaults in the zip file paste it there. Please refer http://old.nabble.com/Re%3A-inconsistency-between-4.4-cvs-and-4.5-for-default-install-directory-of-x11-resources-(Gnuplot)-p27672453.html Regards Tatsuro -------------------------------------- VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] http://pr.mail.yahoo.co.jp/olympic/ |
|
From: sfeam (E. Merritt) <eam...@gm...> - 2010-02-22 02:38:16
|
On Sunday 21 February 2010, Tatsuro MATSUOKA wrote:
> Hello Tait and Ethan
>
>
> --- Ethan Merritt wrote:
>
> > On Sunday 21 February 2010, Tait wrote:
> > >
> > > No complaint if you want to do this privately, but for distribution, are
> > > we allowed to link against readline?
> >
> > Good point. I had momentarily forgotten that the windows version would
> > be provided as a binary image rather than as source.
> >
> > > Gnuplot is not GPL-licensed (and
> > > personally I prefer it that way), and as I just became aware the other
> > > day, it's apparently the FSF's position that linking against readline is
> > > NOT allowed from non-GPLv3 applications.
> >
> > I happen to disagree with them about that. I do not believe that linking
> > to an optional external dynamic library by itself creates a derived work,
> > hence the GPL restrictions are not relevant. But it's a fight we don't
> > need to pick, so let's not.
> >
> > Ethan
> >
>
> Thank you for point the licensing issue about the GPL.
>
> That is not related to only to the readline and serioys problem for binary distribution for windows
> related one.
>
> Pango, freetype, and libiconv are under the GPL (but version is 2nd not GPL v3).
It is a little bit different from that.
readline is GPL "General Public License"
pango and libiconv are LGPL "Lesser General Public License"
a.k.a. "Library General Public License"
see http://svn.gnome.org/svn/pango/trunk/README
http://www.gnu.org/software/libiconv/
The extra "L" is important. It means that you are explicitly
allowed to link to it from software using a different license.
freetype has its own license, but it is essentially BSD, which
means there are no restrictions about linking to it. It is also
possible to obtain freetype under a GPL-compatible license if you
need one, but that is not the case for gnuplot.
see http://freetype.sourceforge.net/license.html
So linking to pango, freetype, and libiconv is not a problem.
Ethan
> For wxt terminal, the mingw libraries are linked. And it does not work without mingwm10.dll.
>
> The libiconv perhaps is be able not to use it. Others are hard to remove them.
>
>
> If the binaries link with the GPL software cannot be allowed, we cannot distribute any windows
> binaries with recent and new features (the cairo related terminal and the wxt terminal). The people
> would like to use these feature, they have to build the gnuplot by themselves.
>
> I have read the GPL licence (Sorry I have read the GPL v2 because I cannot figure out the
> corresponding part in the GPL v3). The license of libraries are different from the that of the GPL'ed
> software itself.
>
> The cygwin binaries that shipped on the gnuplot site (and also on my site) seems to have been linked
> with the libreadline.
>
> Perhaps the license issue is very important. I think that more discussions are required but not
> simply stop to link with GPL'ed libraries.
>
> Regards
>
> Tatsuro
>
> --------------------------------------
> VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
> http://pr.mail.yahoo.co.jp/olympic/
>
|
|
From: Tatsuro M. <tma...@ya...> - 2010-02-22 02:15:19
|
Hello --- "sfeam (Ethan Merritt)" wrote: > > It is a little bit different from that. > > readline is GPL "General Public License" > > pango and libiconv are LGPL "Lesser General Public License" > a.k.a. "Library General Public License" > > see http://svn.gnome.org/svn/pango/trunk/README > http://www.gnu.org/software/libiconv/ > > The extra "L" is important. It means that you are explicitly > allowed to link to it from software using a different license. > > freetype has its own license, but it is essentially BSD, which > means there are no restrictions about linking to it. It is also > possible to obtain freetype under a GPL-compatible license if you > need one, but that is not the case for gnuplot. > > see http://freetype.sourceforge.net/license.html > > > So linking to pango, freetype, and libiconv is not a problem. > > Ethan I really appreciate you for your kind explanations. I have just noticed the difference Lesser General Public License (LGPL) and GNU LIBRARY GENERAL PUBLIC LICENSE. For mingw runtime, http://www.mingw.org/license MinGW runtime: The MinGW base runtime package has been placed in the public domain, and is not governed by copyright. This basically means that you can do what you like with the code. It is no restriction for the copyright. Sorry my confusions. Therefore it is no problem to distribute the binaries in the current style built in the MinGW environments. The readline of cygwin binaries in currently to might be a problem. Anyway I will try the libeditline to avoid the problem. Regards Tatsuro -------------------------------------- VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] http://pr.mail.yahoo.co.jp/olympic/ |
|
From: Tatsuro M. <tma...@ya...> - 2010-02-22 00:57:59
|
Hello For the readline, it is clearly stated in the readline page. *************** Readline is free software, distributed under the terms of the GNU General Public License, version 3. This means that if you want to use Readline in a program that you release or distribute to anyone, the program must be free software and have a GPL-compatible license. If you would like advice on making your license GPL-compatible, contact lic...@gn.... *************** For other GPL'd libraries, I will have looks in more details. Regards Tatsuro --- Tatsuro MATSUOKA wrote: > Hello Tait and Ethan > > > --- Ethan Merritt wrote: > > > On Sunday 21 February 2010, Tait wrote: > > > > > > No complaint if you want to do this privately, but for distribution, are > > > we allowed to link against readline? > > > > Good point. I had momentarily forgotten that the windows version would > > be provided as a binary image rather than as source. > > > > > Gnuplot is not GPL-licensed (and > > > personally I prefer it that way), and as I just became aware the other > > > day, it's apparently the FSF's position that linking against readline is > > > NOT allowed from non-GPLv3 applications. > > > > I happen to disagree with them about that. I do not believe that linking > > to an optional external dynamic library by itself creates a derived work, > > hence the GPL restrictions are not relevant. But it's a fight we don't > > need to pick, so let's not. > > > > Ethan > > > > Thank you for point the licensing issue about the GPL. > > That is not related to only to the readline and serioys problem for binary distribution for > windows > related one. > > Pango, freetype, and libiconv are under the GPL (but version is 2nd not GPL v3). > > For wxt terminal, the mingw libraries are linked. And it does not work without mingwm10.dll. > > The libiconv perhaps is be able not to use it. Others are hard to remove them. > > > If the binaries link with the GPL software cannot be allowed, we cannot distribute any windows > binaries with recent and new features (the cairo related terminal and the wxt terminal). The > people > would like to use these feature, they have to build the gnuplot by themselves. > > I have read the GPL licence (Sorry I have read the GPL v2 because I cannot figure out the > corresponding part in the GPL v3). The license of libraries are different from the that of the > GPL'ed > software itself. > > The cygwin binaries that shipped on the gnuplot site (and also on my site) seems to have been > linked > with the libreadline. > > Perhaps the license issue is very important. I think that more discussions are required but not > simply stop to link with GPL'ed libraries. > > Regards > > Tatsuro > > -------------------------------------- > VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] > http://pr.mail.yahoo.co.jp/olympic/ > > ------------------------------------------------------------------------------ > Download Intel¢î Parallel Studio Eval > Try the new software tools for yourself. Speed compiling, find bugs > proactively, and fine-tune applications for parallel performance. > See why Intel Parallel Studio got high marks during beta. > http://p.sf.net/sfu/intel-sw-dev > _______________________________________________ > gnuplot-beta mailing list > gnu...@li... > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta > -------------------------------------- VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] http://pr.mail.yahoo.co.jp/olympic/ |
|
From: Tatsuro M. <tma...@ya...> - 2010-02-22 00:56:57
|
Hello Sorry I am in confusing state. Please forgive me if I made anyone in frustrated state. However, the clearance for copyright is to be important in order to decide to the style binary distributions on windows. Regards Tatsuro --- Tatsuro MATSUOKA wrote: > Hello > > For the readline, it is clearly stated in the readline page. > > *************** > Readline is free software, distributed under the terms of the GNU General Public License, > version 3. > This means that if you want to use Readline in a program that you release or distribute to > anyone, the > program must be free software and have a GPL-compatible license. If you would like advice on > making > your license GPL-compatible, contact lic...@gn.... > *************** > > For other GPL'd libraries, I will have looks in more details. > > Regards > > Tatsuro > > --- Tatsuro MATSUOKA wrote: > > > Hello Tait and Ethan > > > > > > --- Ethan Merritt wrote: > > > > > On Sunday 21 February 2010, Tait wrote: > > > > > > > > No complaint if you want to do this privately, but for distribution, are > > > > we allowed to link against readline? > > > > > > Good point. I had momentarily forgotten that the windows version would > > > be provided as a binary image rather than as source. > > > > > > > Gnuplot is not GPL-licensed (and > > > > personally I prefer it that way), and as I just became aware the other > > > > day, it's apparently the FSF's position that linking against readline is > > > > NOT allowed from non-GPLv3 applications. > > > > > > I happen to disagree with them about that. I do not believe that linking > > > to an optional external dynamic library by itself creates a derived work, > > > hence the GPL restrictions are not relevant. But it's a fight we don't > > > need to pick, so let's not. > > > > > > Ethan > > > > > > > Thank you for point the licensing issue about the GPL. > > > > That is not related to only to the readline and serioys problem for binary distribution for > > windows > > related one. > > > > Pango, freetype, and libiconv are under the GPL (but version is 2nd not GPL v3). > > > > For wxt terminal, the mingw libraries are linked. And it does not work without mingwm10.dll. > > > > The libiconv perhaps is be able not to use it. Others are hard to remove them. > > > > > > If the binaries link with the GPL software cannot be allowed, we cannot distribute any windows > > binaries with recent and new features (the cairo related terminal and the wxt terminal). The > > people > > would like to use these feature, they have to build the gnuplot by themselves. > > > > I have read the GPL licence (Sorry I have read the GPL v2 because I cannot figure out the > > corresponding part in the GPL v3). The license of libraries are different from the that of > the > > GPL'ed > > software itself. > > > > The cygwin binaries that shipped on the gnuplot site (and also on my site) seems to have been > > linked > > with the libreadline. > > > > Perhaps the license issue is very important. I think that more discussions are required but > not > > simply stop to link with GPL'ed libraries. > > > > Regards > > > > Tatsuro > > > > -------------------------------------- > > VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] > > http://pr.mail.yahoo.co.jp/olympic/ > > > > ------------------------------------------------------------------------------ > > Download Intel召�Parallel Studio Eval > > Try the new software tools for yourself. Speed compiling, find bugs > > proactively, and fine-tune applications for parallel performance. > > See why Intel Parallel Studio got high marks during beta. > > http://p.sf.net/sfu/intel-sw-dev > > _______________________________________________ > > gnuplot-beta mailing list > > gnu...@li... > > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta > > > > > -------------------------------------- > VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] > http://pr.mail.yahoo.co.jp/olympic/ > -------------------------------------- VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] http://pr.mail.yahoo.co.jp/olympic/ |
|
From: Tatsuro M. <tma...@ya...> - 2010-02-21 23:56:52
|
Hello Tait and Ethan --- Ethan Merritt wrote: > On Sunday 21 February 2010, Tait wrote: > > > > No complaint if you want to do this privately, but for distribution, are > > we allowed to link against readline? > > Good point. I had momentarily forgotten that the windows version would > be provided as a binary image rather than as source. > > > Gnuplot is not GPL-licensed (and > > personally I prefer it that way), and as I just became aware the other > > day, it's apparently the FSF's position that linking against readline is > > NOT allowed from non-GPLv3 applications. > > I happen to disagree with them about that. I do not believe that linking > to an optional external dynamic library by itself creates a derived work, > hence the GPL restrictions are not relevant. But it's a fight we don't > need to pick, so let's not. > > Ethan > Thank you for point the licensing issue about the GPL. That is not related to only to the readline and serioys problem for binary distribution for windows related one. Pango, freetype, and libiconv are under the GPL (but version is 2nd not GPL v3). For wxt terminal, the mingw libraries are linked. And it does not work without mingwm10.dll. The libiconv perhaps is be able not to use it. Others are hard to remove them. If the binaries link with the GPL software cannot be allowed, we cannot distribute any windows binaries with recent and new features (the cairo related terminal and the wxt terminal). The people would like to use these feature, they have to build the gnuplot by themselves. I have read the GPL licence (Sorry I have read the GPL v2 because I cannot figure out the corresponding part in the GPL v3). The license of libraries are different from the that of the GPL'ed software itself. The cygwin binaries that shipped on the gnuplot site (and also on my site) seems to have been linked with the libreadline. Perhaps the license issue is very important. I think that more discussions are required but not simply stop to link with GPL'ed libraries. Regards Tatsuro -------------------------------------- VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi] http://pr.mail.yahoo.co.jp/olympic/ |
|
From: Ethan M. <merritt@u.washington.edu> - 2010-02-21 16:40:37
|
On Sunday 21 February 2010, Tait wrote: > > No complaint if you want to do this privately, but for distribution, are > we allowed to link against readline? Good point. I had momentarily forgotten that the windows version would be provided as a binary image rather than as source. > Gnuplot is not GPL-licensed (and > personally I prefer it that way), and as I just became aware the other > day, it's apparently the FSF's position that linking against readline is > NOT allowed from non-GPLv3 applications. I happen to disagree with them about that. I do not believe that linking to an optional external dynamic library by itself creates a derived work, hence the GPL restrictions are not relevant. But it's a fight we don't need to pick, so let's not. Ethan > Tait > > > ...I have tried link with libreadline in building > > gnuplot for windows. > |